A number of voice assisted machine control applications have been developed due to recent advancements in speech recognition software. Examples of these applications include input and storage of electronic patient records, assistance in clinical diagnosis, generation and manipulation of medical images, and manipulation of surgical devices such as microscopes.
Voice activated control of surgical instruments for vitreo-retinal surgery, for example, appears to be a good application for voice actuation technology. Intraocular surgery frequently requires simultaneous use of elaborate and sophisticated pieces of equipment by the surgeon, and coordination and integration of such equipment in an operating room with assistants and one or more nurses. A nurse is typically required to manipulate equipment controls in response to a surgeon's commands. Delays can occur when the nurse is busy setting up or adjusting other parameters when the surgeon's command is received. Delays associated with the human interface between a surgeon and a nurse can increase complexity of the surgery, particularly when conditions inside the patient's eye change rapidly and unexpectedly.
U.S. Pat. No. 4,989,253 discloses a voice control system for moving and focusing a microscope. The system is disadvantageous, however, because it employs a discrete speech recognition algorithm. An operator must therefore give simple verbal commands, which are typically only one or two words, and speak them clearly and slowly. During procedures such as intraocular eye surgery, it is difficult to expect a surgeon to speak in the slow, halting manner required for accurate word identification using a discrete speech recognition algorithm.
U.S. Pat. No. 5,303,148 discloses a voice activated image display controller which requires each operator to speak every command in order to create a stored user-specific pronunciation list. The controller compares received digitized audio signals with the list to determine if a spoken word is a valid command. The controller is disadvantageous because it requires operators to undergo time-consuming training in order to use it.
A need exists, therefore, for a voice command and control system for medical equipment which allows a surgeon to render voice commands using natural, conversational speech and which does not require the surgeon to participate in extensive training of the speech recognition engine. Further, a need exists for a voice command and control system which allows a surgeon to effectively handle volatile and quickly changing patient conditions (e.g., intraocular pressure), and to decrease surgery time without compromising the safety of the surgical procedure.